Critiques of PVA

• Models require too much data

  • Models typically result in wide confidence intervals.

• Models can not be “validated”

  • Focus instead on parameter estimation.

• Models are error prone

Previous slide

Next slide

Back to first slide

View graphic version